Summary: | [EFL][WK2] Compile webkit1 and webkit2 by default. | ||
---|---|---|---|
Product: | WebKit | Reporter: | Mariusz Grzegorczyk <mariusz.g> |
Component: | WebKit2 | Assignee: | Nobody <webkit-unassigned> |
Status: | RESOLVED FIXED | ||
Severity: | Normal | CC: | cshu, d-r, g.czajkowski, gyuyoung.kim, rakuco, ryuan.choi, tmpsantos, webkit.review.bot |
Priority: | P2 | ||
Version: | 528+ (Nightly build) | ||
Hardware: | Unspecified | ||
OS: | Linux | ||
Bug Depends on: | |||
Bug Blocks: | 61838 | ||
Attachments: |
Description
Mariusz Grzegorczyk
2012-04-27 02:32:23 PDT
Created attachment 139156 [details]
Make webkit2 EFL's port compilable from Tools/Scripts/build-webkit script
Comment on attachment 139156 [details]
Make webkit2 EFL's port compilable from Tools/Scripts/build-webkit script
Basically, looks good to me. However, can efl webkit2 be built on trunk ?
(In reply to comment #2) > (From update of attachment 139156 [details]) > Basically, looks good to me. However, can efl webkit2 be built on trunk ? If all related patches will be merged, webkit2 can be built on trunk like gtk port. Mariusz, I am seeing the same issue as the EWS on my machine: "-- Configuring done CMake Error in Source/WebKit2/CMakeLists.txt: Cannot find source file "WebCoreArgumentCodersEfl.cpp". Tried extensions .c .C .c++ .cc .cpp .cxx .m .M .mm .h .hh .h++ .hm .hpp .hxx .in .txx" I am looking into it. mariusz, Could you rebase the patch? I think that WebKit/Efl will be compiled in ews now. I think we need to support --only-webkit and --no-webkit2 options for EFL port. For example, GTK port supports this option as below, - http://trac.webkit.org/browser/trunk/Tools/Scripts/build-webkit#L365 If possible, could you support this option as well? (In reply to comment #6) > I think we need to support --only-webkit and --no-webkit2 options for EFL port. > > For example, GTK port supports this option as below, > - http://trac.webkit.org/browser/trunk/Tools/Scripts/build-webkit#L365 > > If possible, could you support this option as well? --no-webkit2 option is used as $noWebKit2 variable, and --only-webkit is as I see platform independent Created attachment 145116 [details]
Make webkit2 EFL's port compilable from Tools/Scripts/build-webkit script
Rebased.
(In reply to comment #8) > Created an attachment (id=145116) [details] > Make webkit2 EFL's port compilable from Tools/Scripts/build-webkit script > > Rebased. Thank you. informal r+ on my side. Looks good to me. Thanks. Comment on attachment 145116 [details] Make webkit2 EFL's port compilable from Tools/Scripts/build-webkit script View in context: https://bugs.webkit.org/attachment.cgi?id=145116&action=review It looks good. You need to rebase your patch, and I don't think the EWS has actually run it, as ENABLE_WEBKIT2 is still OFF in its CMakeCache.txt. > Tools/Scripts/build-webkit:677 > + $cmakeArgs .= ($cmakeArgs ? " " : "") . "-DENABLE_WEBKIT=On"; > + $cmakeArgs .= " -DENABLE_WEBKIT2=On" if !$noWebKit2; Style nit: we generally use "ON" instead of "On". > It looks good. You need to rebase your patch, and I don't think the EWS has actually run it, as ENABLE_WEBKIT2 is still OFF in its CMakeCache.txt.
Hm, I've ran webkit build like on EWS "perl ./Tools/Scripts/build-webkit --release --efl" and both webkit1 and webkit2 were built. In CMakeCache.txt I have "ENABLE_WEBKIT2:UNINITIALIZED=On". Is this right command, or EWS makes something extra?
(In reply to comment #12) EWS is compiling webkit using "Tools/Scripts/build-webkit --efl". If there is no problems when you build WebKit with your patches, I think EWS also won't have problems.(In reply to comment #12) > > It looks good. You need to rebase your patch, and I don't think the EWS has actually run it, as ENABLE_WEBKIT2 is still OFF in its CMakeCache.txt. > > Hm, I've ran webkit build like on EWS "perl ./Tools/Scripts/build-webkit --release --efl" and both webkit1 and webkit2 were built. In CMakeCache.txt I have "ENABLE_WEBKIT2:UNINITIALIZED=On". Is this right command, or EWS makes something extra? If the feature is not defined in WebKitFeatures, UNINITIALIZED is added to CMakeCache.txt. But, I think we don't need to add ENABLE_WEBKIT2 to WebKitFeatures yet. http://trac.webkit.org/browser/trunk/Source/cmake/WebKitFeatures.cmake#L14 I ran webkit1 and webkit2 with this patch. Then, I could get both libewebkit2.so and libewebkit.so. It seems to me there is no problem. Kubo, do you think ENABLE_WEBKIT2 is off yet ? Mariusz, as kubo said, you need to use uppercase for "On". Created attachment 145759 [details]
Make webkit2 EFL's port compilable from Tools/Scripts/build-webkit script
Rebased, make changes pointed by Raphael
LGTM. Comment on attachment 145759 [details]
Make webkit2 EFL's port compilable from Tools/Scripts/build-webkit script
Looks good to me.
LGTM too. Comment on attachment 145759 [details] Make webkit2 EFL's port compilable from Tools/Scripts/build-webkit script Clearing flags on attachment: 145759 Committed r119678: <http://trac.webkit.org/changeset/119678> All reviewed patches have been landed. Closing bug. |